Update changelog.

[youtubedl] / youtube_dl / extractor / vk.py

diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py
index 0805e3c083937d7f58acbe4b872d41c19f8d1c02..cfc5ffd8b02082d1f6d3035ba7b02dc230b8b2ed 100644 (file)
--- a/youtube_dl/extractor/vk.py
+++ b/youtube_dl/extractor/vk.py
@@ -3,19 +3,18 @@ from __future__ import unicode_literals
 
 import re
 import json
 
 import re
 import json

+import sys

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

-from ..compat import (
-    compat_str,
-    compat_urllib_parse,
-)
+from ..compat import compat_str

 from ..utils import (
     ExtractorError,
 from ..utils import (
     ExtractorError,

+    int_or_none,

     orderedSet,
     orderedSet,

-    sanitized_Request,

     str_to_int,
     unescapeHTML,
     unified_strdate,
     str_to_int,
     unescapeHTML,
     unified_strdate,

+    urlencode_postdata,

 )
 from .vimeo import VimeoIE
 from .pladform import PladformIE
 )
 from .vimeo import VimeoIE
 from .pladform import PladformIE


@@ -27,12 +26,16 @@ class VKIE(InfoExtractor):
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
     _VALID_URL = r'''(?x)
                     https?://
                         (?:

-                            (?:m\.)?vk\.com/video_ext\.php\?.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+)|
+                            (?:
+                                (?:m\.)?vk\.com/video_|
+                                (?:www\.)?daxab.com/
+                            )
+                            ext\.php\?(?P<embed_query>.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+).*)|

                             (?:
                                 (?:m\.)?vk\.com/(?:.+?\?.*?z=)?video|
                             (?:
                                 (?:m\.)?vk\.com/(?:.+?\?.*?z=)?video|

-                                (?:www\.)?biqle\.ru/watch/
+                                (?:www\.)?daxab.com/embed/

                             )
                             )

-                            (?P<videoid>[^s].*?)(?:\?(?:.*\blist=(?P<list_id>[\da-f]+))?|%2F|$)
+                            (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>[\da-f]+))?

                         )
                     '''
     _NETRC_MACHINE = 'vk'
                         )
                     '''
     _NETRC_MACHINE = 'vk'


@@ -76,7 +79,8 @@ class VKIE(InfoExtractor):
                 'duration': 101,
                 'upload_date': '20120730',
                 'view_count': int,
                 'duration': 101,
                 'upload_date': '20120730',
                 'view_count': int,

-            }
+            },
+            'skip': 'This video has been removed from public access.',

         },
         {
             # VIDEO NOW REMOVED
         },
         {
             # VIDEO NOW REMOVED


@@ -141,16 +145,29 @@ class VKIE(InfoExtractor):
             'url': 'https://vk.com/video276849682_170681728',
             'info_dict': {
                 'id': 'V3K4mi0SYkc',
             'url': 'https://vk.com/video276849682_170681728',
             'info_dict': {
                 'id': 'V3K4mi0SYkc',

-                'ext': 'mp4',
+                'ext': 'webm',

                 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
                 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",

-                'description': 'md5:bf9c26cfa4acdfb146362682edd3827a',
-                'duration': 179,
+                'description': 'md5:d9903938abdc74c738af77f527ca0596',
+                'duration': 178,

                 'upload_date': '20130116',
                 'uploader': "Children's Joy Foundation",
                 'uploader_id': 'thecjf',
                 'view_count': int,
             },
         },
                 'upload_date': '20130116',
                 'uploader': "Children's Joy Foundation",
                 'uploader_id': 'thecjf',
                 'view_count': int,
             },
         },

+        {
+            # video key is extra_data not url\d+
+            'url': 'http://vk.com/video-110305615_171782105',
+            'md5': 'e13fcda136f99764872e739d13fac1d1',
+            'info_dict': {
+                'id': '171782105',
+                'ext': 'mp4',
+                'title': 'S-Dance, репетиции к The way show',
+                'uploader': 'THE WAY SHOW | 17 апреля',
+                'upload_date': '20160207',
+                'view_count': int,
+            },
+        },

         {
             # removed video, just testing that we match the pattern
             'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
         {
             # removed video, just testing that we match the pattern
             'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',


@@ -161,11 +178,6 @@ class VKIE(InfoExtractor):
             'url': 'https://vk.com/video205387401_164765225',
             'only_matching': True,
         },
             'url': 'https://vk.com/video205387401_164765225',
             'only_matching': True,
         },

-        {
-            # vk wrapper
-            'url': 'http://www.biqle.ru/watch/847655_160197695',
-            'only_matching': True,
-        },

         {
             # pladform embed
             'url': 'https://vk.com/video-76116461_171554880',
         {
             # pladform embed
             'url': 'https://vk.com/video-76116461_171554880',


@@ -178,7 +190,7 @@ class VKIE(InfoExtractor):
         if username is None:
             return
 
         if username is None:
             return
 

-        login_page = self._download_webpage(
+        login_page, url_handle = self._download_webpage_handle(

             'https://vk.com', None, 'Downloading login page')
 
         login_form = self._hidden_inputs(login_page)
             'https://vk.com', None, 'Downloading login page')
 
         login_form = self._hidden_inputs(login_page)


@@ -188,11 +200,26 @@ class VKIE(InfoExtractor):
             'pass': password.encode('cp1251'),
         })
 
             'pass': password.encode('cp1251'),
         })
 

-        request = sanitized_Request(
-            'https://login.vk.com/?act=login',
-            compat_urllib_parse.urlencode(login_form).encode('utf-8'))
+        # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header
+        # and expects the first one to be set rather than second (see
+        # https://github.com/rg3/youtube-dl/issues/9841#issuecomment-227871201).
+        # As of RFC6265 the newer one cookie should be set into cookie store
+        # what actually happens.
+        # We will workaround this VK issue by resetting the remixlhk cookie to
+        # the first one manually.
+        cookies = url_handle.headers.get('Set-Cookie')
+        if sys.version_info[0] >= 3:
+            cookies = cookies.encode('iso-8859-1')
+        cookies = cookies.decode('utf-8')
+        remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies)
+        if remixlhk:
+            value, domain = remixlhk.groups()
+            self._set_cookie(domain, 'remixlhk', value)
+

         login_page = self._download_webpage(
         login_page = self._download_webpage(

-            request, None, note='Logging in as %s' % username)
+            'https://login.vk.com/?act=login', None,
+            note='Logging in as %s' % username,
+            data=urlencode_postdata(login_form))

 
         if re.search(r'onLoginFailed', login_page):
             raise ExtractorError(
 
         if re.search(r'onLoginFailed', login_page):
             raise ExtractorError(


@@ -205,20 +232,21 @@ class VKIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('videoid')
 
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('videoid')
 

-        if not video_id:
+        if video_id:
+            info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
+            # Some videos (removed?) can only be downloaded with list id specified
+            list_id = mobj.group('list_id')
+            if list_id:
+                info_url += '&list=%s' % list_id
+        else:
+            info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query')

             video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
 
             video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
 

-        info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
-
-        # Some videos (removed?) can only be downloaded with list id specified
-        list_id = mobj.group('list_id')
-        if list_id:
-            info_url += '&list=%s' % list_id
-

         info_page = self._download_webpage(info_url, video_id)
 
         error_message = self._html_search_regex(
         info_page = self._download_webpage(info_url, video_id)
 
         error_message = self._html_search_regex(

-            r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
+            [r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
+                r'(?s)<div[^>]+id="video_ext_msg"[^>]*>(.+?)</div>'],

             info_page, 'error message', default=None)
         if error_message:
             raise ExtractorError(error_message, expected=True)
             info_page, 'error message', default=None)
         if error_message:
             raise ExtractorError(error_message, expected=True)


@@ -293,17 +321,22 @@ class VKIE(InfoExtractor):
         view_count = None
         views = self._html_search_regex(
             r'"mv_views_count_number"[^>]*>(.+?\bviews?)<',
         view_count = None
         views = self._html_search_regex(
             r'"mv_views_count_number"[^>]*>(.+?\bviews?)<',

-            info_page, 'view count', fatal=False)
+            info_page, 'view count', default=None)

         if views:
             view_count = str_to_int(self._search_regex(
                 r'([\d,.]+)', views, 'view count', fatal=False))
 
         if views:
             view_count = str_to_int(self._search_regex(
                 r'([\d,.]+)', views, 'view count', fatal=False))
 

-        formats = [{
-            'format_id': k,
-            'url': v,
-            'width': int(k[len('url'):]),
-        } for k, v in data.items()
-            if k.startswith('url')]
+        formats = []
+        for k, v in data.items():
+            if not k.startswith('url') and not k.startswith('cache') and k != 'extra_data' or not v:
+                continue
+            height = int_or_none(self._search_regex(
+                r'^(?:url|cache)(\d+)', k, 'height', default=None))
+            formats.append({
+                'format_id': k,
+                'url': v,
+                'height': height,
+            })

         self._sort_formats(formats)
 
         return {
         self._sort_formats(formats)
 
         return {